The CQC Single Assessment Framework in plain English

The Single Assessment Framework is CQC's replacement for the older inspection approach that ran for the previous decade. Three things to understand about the structure.

The five key questions are preserved. Safe, Effective, Caring, Responsive, Well-led. These are the headings any rating sits under. Anyone familiar with the old framework will recognise these immediately.

Quality statements replace key lines of enquiry. The old framework had Key Lines of Enquiry (KLOEs) under each key question. The new framework has Quality Statements. There are roughly 30 of them in total, distributed across the five key questions, and they are phrased from the provider's perspective as "we statements" ("We work in partnership with people"), with paired "I statements" expressing the experience the same quality is meant to produce for the person using the service ("I have the information I need"). Read the framework directly for the exact wording per statement.

Evidence categories tell the inspector what to look at. Under each quality statement, the framework lists categories of evidence the inspector will consider. There are six categories: People's experience, Feedback from staff and leaders, Observation, Processes, Outcomes, and Feedback from partners. Not every category applies to every quality statement; the framework specifies which apply where.

Put together: an inspection assesses each quality statement by drawing on the relevant evidence categories, and the quality statements roll up into a rating against each of the five key questions, which rolls up into an overall rating.

Three things changed that materially affect what providers should be doing day to day.

Evidence is more explicit.The old KLOE approach was prompt-shaped ("How does the service know that care is safe?"); the new framework is evidence-shaped (the inspector will look at specific categories of evidence to make the call). This sounds like an aesthetic change. It is not. It changes what providers should be recording. Under the old approach, you could plausibly answer a KLOE in a conversation. Under the new approach, the inspector is looking for evidence, not narrative.

Continuous assessment is now the default. The framework is designed to support ongoing assessment between inspections, not a single point-in-time judgment. CQC can update ratings outside of a full inspection if evidence accumulates that warrants it. This makes the evidence trail you keep in normal operations more directly important, because it can feed into a rating change without anyone scheduling an inspection first.

Outcomes get more weight.The old framework leaned toward processes ("do you have a policy for X?"). The new framework adds explicit categories around outcomes and people's experience. A service can no longer be rated well purely on the strength of its policy folder; the framework now asks what happened to the people in your care.

Important for context, because consultancies sometimes oversell the change.

The five key questions are unchanged.Safe, Effective, Caring, Responsive, Well-led mean roughly what they meant before. If you have an old inspection report against the previous framework, it is still a reasonable starting point for understanding where the regulator's attention sits.

The rating ladder is unchanged. Outstanding, Good, Requires Improvement, Inadequate. Same four levels.

The notification and duty-of-candour regulations are unchanged. Regulation 18 and Regulation 20 are statutory duties that sit outside the framework; the framework changed but those did not.

The inspection mechanics are similar. Announced inspections, focused inspections, comprehensive inspections, off-site evidence review, on-site fieldwork. The shape of how an inspection is run is broadly the same; the analytical framework underneath it changed.

This is the practical bit. For each evidence category, what does it mean for what you actually do.

People's experience. Feedback from people using the service. Patient or service-user surveys, complaints data, comments captured during care, family feedback. The framework wants to see that you are systematically capturing this, not just responding when something goes wrong. Daily work: a habit of recording feedback during the work, not waiting for someone to write a formal complaint.

Feedback from staff and leaders. Staff surveys, staff one-to-ones, exit interviews, supervision records. The framework treats how staff feel about the service as evidence about how the service is run. Daily work: supervision records that contain real content, not template-tick. Exit interviews that get filed.

Observation. What the inspector sees on site. The atmosphere in waiting areas, how staff interact with patients, whether the environment matches the records. This category cannot be evidenced in advance; it can only be undermined in advance (a chaotic site visit reads worse than calm records do). Daily work: keep the actual service consistent with the recorded service.

Processes. The things you document because the work requires it. Incident records, complaint records, governance meeting minutes, training matrix, policies, risk register. This is the category that maps most cleanly to the old framework. Daily work: keep the records honest and current.

Outcomes. What happened to the people in your care. Clinical outcomes where you measure them, complications, re-admissions, complaints resolution rates, safeguarding case outcomes. This category is harder for small independent services that do not run formal outcome studies; the framework expects you to use what data you have. Daily work: track the small number of outcome metrics that matter for your service and review them periodically.

Feedback from partners. What other organisations say about you. Referring GPs, contracted commissioners, NHS partners, sector regulators, professional bodies. The framework will draw on this category where partners exist. Daily work: keep relationships with your partners honest, and notice when feedback drifts.

Three recommendations from inside the regulator.

Read the quality statements at the start of every quarter. They are short. CQC publishes them. They tell you exactly what the inspector is assessing. If you read them once a quarter and ask "what would the evidence look like for this", you are doing the work the framework is asking for.

Map your current evidence to the six categories. Take an A4 page. Down the left, list the six categories. For each, write the records or processes you currently have that would feed that category. The gaps are your roadmap.

Treat people's experience and outcomes as separate from processes. These are the two categories most providers under-evidence, because they are the two that did not exist as explicit categories under the old framework. If your evidence is heavy on processes and thin on experience and outcomes, you are running an old-framework service inside a new framework, and the rating will reflect that.

CQC publishes the framework directly: the quality statements, the evidence categories, and the assessment methodology. Read the published material rather than any consultancy paraphrase. The framework is the single most useful thing to read at the start of every inspection cycle.